EXPERIMENTAL MANIPULATION OF NATURAL PLANT-COMMUNITIES

Attempts to elucidate the factors controlling the structure of plant c ommunities have relied increasingly on field experiments. This is a po werful approach for testing theoretical predictions that offers import ant advantages over observational and comparative studies. However, fi eld experiments suffer from intrinsic difficulties as well as more-eas ily remediable limitations. Recent progress has been made by new appro aches including the use of multifactor experiments, and the developmen t and dissemination of better statistical tools.